Latest / Hardware Startups with Fexingo: Devices, Robotics, and Manufacturing Tech Companies / How Hardware Startups Use Solder Paste Stencils for Consistency
Transcript
- Lucas: You can have the most brilliant circuit design in the world, but if the solder paste isn't deposited consistently, your board is a coin flip. And the single most important tool for consistent paste deposition is something most hardware startups don't think about until it ruins a build: the solder paste stencil. Luna: The metal sheet with all the little holes. I've definitely seen teams treat it as a commodity — just order the cheapest one from the first supplier they find. Lucas: Right, and that's exactly the problem. A stencil isn't a commodity. It's a precision tool. We're talking about a stainless steel or nickel foil, typically one hundred to two hundred microns thick, with apertures laser-cut or electroformed to match every pad on your board. The geometry of those apertures — the shape, the wall angle, the surface finish — directly controls how much paste lands on each pad, and that determines everything from solder joint strength to the risk of defects like tombstoning or solder balling. Luna: And the tolerance is measured in microns, right? A few microns off on aperture size can change paste volume by ten or fifteen percent. Lucas: Exactly. So let me give you a concrete example. I was talking to a medical-device startup that was building a wearable sensor. They were seeing tombstoning rates of about twelve percent on a specific 0402 capacitor. For a medical device, that's catastrophic. They'd redesigned the pad geometry twice, tweaked the reflow profile, nothing worked. Finally they looked at the stencil. They were using an electroformed stencil with a nominal thickness of one hundred fifty microns, but the apertures for that capacitor had an area ratio of about 0.5. The rule of thumb in stencil printing is that you need an area ratio above 0.66 for reliable paste release. Below that, the paste sticks to the stencil walls instead of transferring to the pad. Luna: So the paste wasn't releasing evenly, which meant some pads got too little paste and the component lifted during reflow. Classic tombstoning mechanism. Lucas: Exactly. They switched to a laser-cut stencil with a tapered wall angle — about five degrees — which effectively increased the area ratio for those small apertures to 0.7. Tombstoning dropped to under one percent overnight. The stencil cost them about three hundred dollars more than the electroformed one, but they saved thousands in rework and delayed shipments. Luna: That's a great case study. But for a startup that's only building fifty prototypes, is it really worth spending extra on a premium stencil? I've seen teams argue that the yield hit is acceptable at low volumes. Lucas: It depends on what you're prototyping. If those fifty boards are headed to clinical trials or beta customers, a twelve percent defect rate is not acceptable — you lose credibility. And even for internal testing, defective boards cost time. You have to debug whether the issue is the design or the assembly process. So I'd argue it's almost always worth investing in a good stencil, even for low volumes. The cost difference is tiny relative to the engineering time you burn on false failures. Luna: So what should a startup specify when ordering a stencil? What are the key parameters beyond area ratio? Lucas: First, thickness. A good starting point is one hundred twenty-five microns for standard pitch components, but if you have a mix of fine-pitch QFPs and large power pads, you might need a step stencil — thinner in some areas, thicker in others. Step stencils are more expensive but can solve the problem of too much paste on fine-pitch leads and too little on ground planes. Second, aperture design. For fine-pitch, you want a slight taper — the aperture should be slightly larger on the squeegee side than on the board side, so the paste releases cleanly. Third, surface finish. Electropolished or nanocoated stencils reduce paste sticking and require less frequent cleaning. Fourth, the frame tension. A loose stencil causes gasketing issues and paste smearing. Luna: And cleaning frequency? I've seen teams run a hundred boards without cleaning the stencil and wonder why defects appear halfway through the run. Lucas: That's the other hidden variable. Paste dries in the apertures over time, especially in low-humidity environments. For a typical print run, you should clean the stencil every ten to fifteen prints — or more often if you're using a fast-evaporating flux. Under-stencil cleaning with a dry or solvent wipe between prints is standard in high-volume production, but for startups it's often done manually. The key is consistency: if you clean every tenth board, the paste volume stays stable. If you clean irregularly, you get drift. Luna: I've also heard about using nickel instead of stainless steel for stencils. Is that worth it for a startup? Lucas: Nickel stencils are typically electroformed, which gives very smooth aperture walls — great for ultra fine pitch components like 0.3 millimeter BGAs. But they're more expensive and can be less durable than laser-cut stainless. For most startup applications, laser-cut stainless with a good surface finish is the sweet spot. I'd only recommend nickel if you're consistently working with pitch below 0.4 millimeters or if you need extremely consistent paste volume for RF or medical applications. Luna: So the takeaway is: don't treat the stencil as an afterthought. It's a process parameter, like reflow temperature or solder alloy. You need to design it for your specific board. Lucas: Exactly. And if you're a hardware founder, I'd suggest ordering two stencils for your first production run — one with your best-guess aperture design, and one with slightly modified apertures for critical components. The cost of a second stencil is trivial compared to a respin. Luna: That's a practical tip. And while we're on the topic of practical investments, I want to note something about this show. We deliberately don't run ads. No sponsors, no affiliate links. The reason is simple: we want these conversations to be useful for what you're building, without any commercial filter. Lucas: Yeah, and we've kept it that way from episode one. If you find value in the specifics — like that area ratio rule or the step stencil trick — the only way this show stays sustainable is if listeners who can, support it. There's a link at buy me a coffee dot com slash fexingo. Luna: It's not a subscription, not a paywall. Just a way for people who get something out of the show to keep it ad-free and focused on the technical details that matter. Lucas: Right. So back to stencils: one more thing I want to mention is the role of the squeegee. The material and angle of the squeegee blade also affect paste deposition. Metal squeegees are standard now — they last longer and provide better paste roll than polyurethane. But the angle matters: a sixty-degree attack angle is typical for standard pastes, but for finer pitches, some shops use a forty-five-degree angle to improve paste filling. Luna: And print speed? Too fast and the paste doesn't fill the apertures, too slow and it can smear. Lucas: Print speed is usually between twenty and fifty millimeters per second. Slower for fine-pitch, faster for large pads. But the best advice I can give a startup is to run a simple DOE — design of experiments — on your stencil printer. Vary speed, pressure, and separation speed, and measure paste height with a solder paste inspection system. Even a basic SPI system can pay for itself in one build by catching problems early. Luna: So to summarize: the stencil is not a commodity. Design it with area ratio above 0.66, choose laser-cut stainless for most applications, consider step stencils for mixed components, and clean it regularly. That's the recipe for consistent paste deposition. Lucas: And if you're a startup, that consistency translates directly into higher yield, fewer debugging cycles, and faster time to market. The stencil is a small line item with outsized impact. Luna: Next time we'll look at a related topic: how to choose between no-clean and water-soluble solder paste for prototyping. There's a trade-off in reliability and cleaning cost that catches a lot of teams off guard. Lucas: That's a great one. Until then, don't let your stencil be the weak link.